To address the great need for improved therapies for adult glioma, this project will evaluate molecular features of tumor tissue, serologic immune factors and genetic polymorphisms as biomarkers of patient surival. The proposed studies are additions to ongoing recruitment of patients in a 16 year R01 funded population-based case control study and 5 years of SPORE funded clinic-based research in the UCSF Neuro-oncology Clinic (NOC). Together these sources provide among the country s largest datasets of well-characterized adult glioma patients with polymorphism, serologic, tumor marker, demographic and other epidemiologic, treatment, and survival data. We will continue to determine vital status and relevant treatment information for population based adult onset glioma cases diagnosed 1997-99 and 2001-04 and patients accrued through the UCSF NOC 2002-2006 (total number ~1500) and (b) accrue ~720 patients (questionnaire, blood, buccal, and tumor specimens) in the UCSF NOC 2007-2011. (Another population based series to begin May 2006 will bring the total number of patients diagnosed from 2006-2011 to ~960). Several biomarkers identified in our ongoing and other studies warrant further research. In Aims 1,2 of this proposal we will obtain a greater understanding of the mechanisms of improved survival among glioblastoma cases with elevated versus normal or borderline serum IgE levels observed in our current series and validate and expand findings to other potentially related serum markers (sCD23 and sCD14), tumor markers (CD23 protein and IL13RA2 mRNA expression), and constitutive SNPs in IL4, IL13, IL4R, IL13RA1, and IL13RA2. These markers will also be assessed in relation to tumor TP53 mutation and expression and EGFR amplification and expression (markers measured in the parent R01 study). In Aim 3 we will determine whether polymorphisms in MGMT or tumor TP53 mutation or expression influence survival in the presence or absence of tumor MGMT methylation in patients treated or not treated with temozolomide. In Aim 4 we will validate promising markers obtained from Aims 2 and 3 in newly diagnosed patients seen at the UCSF NOC from July 1, 2007-June 30, 2011 in ~120 of these GM cases on clinical trial protocols. Identification of predictive biomarkers related glioma survival is likely to aid in choosing the best treatments for each glioma patient, in enhancing definition of homogeneous subgroups for clinical trials based on uniform prognosis for rapid treatment evaluation, and in providing better prognostic information to patients.